… | |
… | |
18 | vendors) - usually there is only one. |
18 | vendors) - usually there is only one. |
19 | |
19 | |
20 | Each platform gives you access to a number of OpenCL::Device objects, e.g. |
20 | Each platform gives you access to a number of OpenCL::Device objects, e.g. |
21 | your graphics card. |
21 | your graphics card. |
22 | |
22 | |
23 | From a platform and some devices, you create an OpenCL::Context, which is |
23 | From a platform and some device(s), you create an OpenCL::Context, which is |
24 | a very central object in OpenCL: Once you have a context you can create |
24 | a very central object in OpenCL: Once you have a context you can create |
25 | most other objects: |
25 | most other objects: |
26 | |
26 | |
27 | OpenCL::Program objects, which store source code and, after building |
27 | OpenCL::Program objects, which store source code and, after building for a |
28 | ("compiling and linking"), also binary programs. For each kernel function |
28 | specific device ("compiling and linking"), also binary programs. For each |
29 | in a program you can then create an OpenCL::Kernel object which represents |
29 | kernel function in a program you can then create an OpenCL::Kernel object |
30 | basically a function call with argument values. |
30 | which represents basically a function call with argument values. |
31 | |
31 | |
32 | OpenCL::Memory objects of various flavours: OpenCL::Buffers objects (flat |
32 | OpenCL::Memory objects of various flavours: OpenCL::Buffers objects (flat |
33 | memory areas, think array) and OpenCL::Image objects (think 2d or 3d |
33 | memory areas, think arrays or structs) and OpenCL::Image objects (think 2d |
34 | array) for bulk data and input and output for kernels. |
34 | or 3d array) for bulk data and input and output for kernels. |
35 | |
35 | |
36 | OpenCL::Sampler objects, which are kind of like texture filter modes in |
36 | OpenCL::Sampler objects, which are kind of like texture filter modes in |
37 | OpenGL. |
37 | OpenGL. |
38 | |
38 | |
39 | OpenCL::Queue objects - command queues, which allow you to submit memory |
39 | OpenCL::Queue objects - command queues, which allow you to submit memory |
… | |
… | |
52 | |
52 | |
53 | OpenCL manpages: |
53 | OpenCL manpages: |
54 | |
54 | |
55 | http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/ |
55 | http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/ |
56 | |
56 | |
|
|
57 | Here's a tutorial from AMD (very AMD-centric, too), not sure how useful it |
|
|
58 | is, but at least it's free of charge: |
|
|
59 | |
|
|
60 | http://developer.amd.com/zones/OpenCLZone/courses/Documents/Introduction_to_OpenCL_Programming%20Training_Guide%20%28201005%29.pdf |
|
|
61 | |
|
|
62 | If you are into UML class diagrams, the following diagram might help - if |
|
|
63 | not, it will be mildly cofusing: |
|
|
64 | |
|
|
65 | http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/classDiagram.html |
|
|
66 | |
57 | =head1 BASIC WORKFLOW |
67 | =head1 BASIC WORKFLOW |
58 | |
68 | |
59 | To get something done, you basically have to do this once: |
69 | To get something done, you basically have to do this once (refer to the |
|
|
70 | examples below for actual code, this is just a high-level description): |
60 | |
71 | |
61 | Find some platform (e.g. the first one) and some device (e.g. the first |
72 | Find some platform (e.g. the first one) and some device(s) (e.g. the first |
62 | device you can find), and create a context from those. |
73 | device of the platform), and create a context from those. |
63 | |
74 | |
64 | Create a command queue from your context, and program objects from your |
75 | Create program objects from your OpenCL source code, then build (compile) |
65 | OpenCL source code, build the programs. |
76 | the programs for each device you want to run them on. |
66 | |
77 | |
67 | Create kernel objects for all kernels you want to use. |
78 | Create kernel objects for all kernels you want to use (surprisingly, these |
|
|
79 | are not device-specific). |
68 | |
80 | |
69 | Then, to execute stuff, you repeat this: |
81 | Then, to execute stuff, you repeat these steps, possibly resuing or |
|
|
82 | sharing some buffers: |
70 | |
83 | |
71 | Create some input and output buffers from your context. Initialise the |
84 | Create some input and output buffers from your context. Set these as |
72 | input buffers with data. Set these as arguments to your kernel. |
85 | arguments to your kernel. |
|
|
86 | |
|
|
87 | Enqueue buffer writes to initialise your input buffers (when not |
|
|
88 | initialised at creation time). |
73 | |
89 | |
74 | Enqueue the kernel execution. |
90 | Enqueue the kernel execution. |
75 | |
91 | |
76 | Enqueue buffer reads for your output buffer to read results. |
92 | Enqueue buffer reads for your output buffer to read results. |
77 | |
93 | |
78 | The next section shows how this can be done. |
|
|
79 | |
|
|
80 | =head1 EXAMPLES |
94 | =head1 EXAMPLES |
81 | |
95 | |
82 | =head2 Enumerate all devices and get contexts for them. |
96 | =head2 Enumerate all devices and get contexts for them. |
83 | |
97 | |
|
|
98 | Best run this once to get a feel for the platforms and devices in your |
|
|
99 | system. |
|
|
100 | |
84 | for my $platform (OpenCL::platforms) { |
101 | for my $platform (OpenCL::platforms) { |
85 | warn $platform->info (OpenCL::PLATFORM_NAME); |
102 | printf "platform: %s\n", $platform->info (OpenCL::PLATFORM_NAME); |
86 | warn $platform->info (OpenCL::PLATFORM_EXTENSIONS); |
103 | printf "extensions: %s\n", $platform->info (OpenCL::PLATFORM_EXTENSIONS); |
87 | for my $device ($platform->devices) { |
104 | for my $device ($platform->devices) { |
88 | warn $device->info (OpenCL::DEVICE_NAME); |
105 | printf "+ device: %s\n", $device->info (OpenCL::DEVICE_NAME); |
89 | my $ctx = $device->context_simple; |
106 | my $ctx = $device->context; |
90 | # do stuff |
107 | # do stuff |
91 | } |
108 | } |
92 | } |
109 | } |
93 | |
110 | |
94 | =head2 Get a useful context and a command queue. |
111 | =head2 Get a useful context and a command queue. |
95 | |
112 | |
96 | my $dev = ((OpenCL::platforms)[0]->devices)[0]; |
113 | This is a useful boilerplate for any OpenCL program that only wants to use |
97 | my $ctx = $dev->context_simple; |
114 | one device, |
98 | my $queue = $ctx->queue ($dev); |
115 | |
|
|
116 | my ($platform) = OpenCL::platforms; # find first platform |
|
|
117 | my ($dev) = $platform->devices; # find first device of platform |
|
|
118 | my $ctx = $platform->context (undef, [$dev]); # create context out of those |
|
|
119 | my $queue = $ctx->queue ($dev); # create a command queue for the device |
99 | |
120 | |
100 | =head2 Print all supported image formats of a context. |
121 | =head2 Print all supported image formats of a context. |
101 | |
122 | |
|
|
123 | Best run this once for your context, to see whats available and how to |
|
|
124 | gather information. |
|
|
125 | |
102 | for my $type (OpenCL::MEM_OBJECT_IMAGE2D, OpenCL::MEM_OBJECT_IMAGE3D) { |
126 | for my $type (OpenCL::MEM_OBJECT_IMAGE2D, OpenCL::MEM_OBJECT_IMAGE3D) { |
103 | say "supported image formats for ", OpenCL::enum2str $type; |
127 | print "supported image formats for ", OpenCL::enum2str $type, "\n"; |
104 | |
128 | |
105 | for my $f ($ctx->supported_image_formats (0, $type)) { |
129 | for my $f ($ctx->supported_image_formats (0, $type)) { |
106 | printf " %-10s %-20s\n", OpenCL::enum2str $f->[0], OpenCL::enum2str $f->[1]; |
130 | printf " %-10s %-20s\n", OpenCL::enum2str $f->[0], OpenCL::enum2str $f->[1]; |
107 | } |
131 | } |
108 | } |
132 | } |
… | |
… | |
111 | then asynchronously. |
135 | then asynchronously. |
112 | |
136 | |
113 | my $buf = $ctx->buffer_sv (OpenCL::MEM_COPY_HOST_PTR, "helmut"); |
137 | my $buf = $ctx->buffer_sv (OpenCL::MEM_COPY_HOST_PTR, "helmut"); |
114 | |
138 | |
115 | $queue->enqueue_read_buffer ($buf, 1, 1, 3, my $data); |
139 | $queue->enqueue_read_buffer ($buf, 1, 1, 3, my $data); |
116 | warn $data; |
140 | print "$data\n"; |
117 | |
141 | |
118 | my $ev = $queue->enqueue_read_buffer ($buf, 0, 1, 3, my $data); |
142 | my $ev = $queue->enqueue_read_buffer ($buf, 0, 1, 3, my $data); |
119 | $ev->wait; |
143 | $ev->wait; |
120 | warn $data; |
144 | print "$data\n"; # prints "elm" |
121 | |
145 | |
122 | =head2 Create and build a program, then create a kernel out of one of its |
146 | =head2 Create and build a program, then create a kernel out of one of its |
123 | functions. |
147 | functions. |
124 | |
148 | |
125 | my $src = ' |
149 | my $src = ' |
126 | __kernel void |
150 | __kernel void |
127 | squareit (__global float *input, __global float *output) |
151 | squareit (__global float *input, __global float *output) |
128 | { |
152 | { |
129 | size_t id = get_global_id (0); |
153 | $id = get_global_id (0); |
130 | output [id] = input [id] * input [id]; |
154 | output [id] = input [id] * input [id]; |
131 | } |
155 | } |
132 | '; |
156 | '; |
133 | |
157 | |
134 | my $prog = $ctx->program_with_source ($src); |
158 | my $prog = $ctx->program_with_source ($src); |
135 | |
159 | |
|
|
160 | # build croaks on compile errors, so catch it and print the compile errors |
136 | eval { $prog->build ($dev); 1 } |
161 | eval { $prog->build ($dev); 1 } |
137 | or die $prog->build_info ($dev, OpenCL::PROGRAM_BUILD_LOG); |
162 | or die $prog->build_info ($dev, OpenCL::PROGRAM_BUILD_LOG); |
138 | |
163 | |
139 | my $kernel = $prog->kernel ("squareit"); |
164 | my $kernel = $prog->kernel ("squareit"); |
140 | |
165 | |
141 | =head2 Create some input and output float buffers, then call squareit on them. |
166 | =head2 Create some input and output float buffers, then call the |
|
|
167 | 'squareit' kernel on them. |
142 | |
168 | |
143 | my $input = $ctx->buffer_sv (OpenCL::MEM_COPY_HOST_PTR, pack "f*", 1, 2, 3, 4.5); |
169 | my $input = $ctx->buffer_sv (OpenCL::MEM_COPY_HOST_PTR, pack "f*", 1, 2, 3, 4.5); |
144 | my $output = $ctx->buffer (0, OpenCL::SIZEOF_FLOAT * 5); |
170 | my $output = $ctx->buffer (0, OpenCL::SIZEOF_FLOAT * 5); |
145 | |
171 | |
146 | # set buffer |
172 | # set buffer |
… | |
… | |
152 | |
178 | |
153 | # enqueue a synchronous read |
179 | # enqueue a synchronous read |
154 | $queue->enqueue_read_buffer ($output, 1, 0, OpenCL::SIZEOF_FLOAT * 4, my $data); |
180 | $queue->enqueue_read_buffer ($output, 1, 0, OpenCL::SIZEOF_FLOAT * 4, my $data); |
155 | |
181 | |
156 | # print the results: |
182 | # print the results: |
157 | say join ", ", unpack "f*", $data; |
183 | printf "%s\n", join ", ", unpack "f*", $data; |
158 | |
184 | |
159 | =head2 The same enqueue operations as before, but assuming an out-of-order queue, |
185 | =head2 The same enqueue operations as before, but assuming an out-of-order queue, |
160 | showing off barriers. |
186 | showing off barriers. |
161 | |
187 | |
162 | # execute it for all 4 numbers |
188 | # execute it for all 4 numbers |
… | |
… | |
185 | |
211 | |
186 | =head1 DOCUMENTATION |
212 | =head1 DOCUMENTATION |
187 | |
213 | |
188 | =head2 BASIC CONVENTIONS |
214 | =head2 BASIC CONVENTIONS |
189 | |
215 | |
190 | This is not a 1:1 C-style translation of OpenCL to Perl - instead I |
216 | This is not a one-to-one C-style translation of OpenCL to Perl - instead |
191 | attempted to make the interface as type-safe as possible and introducing |
217 | I attempted to make the interface as type-safe as possible by introducing |
192 | object syntax where it makes sense. There are a number of important |
218 | object syntax where it makes sense. There are a number of important |
193 | differences between the OpenCL C API and this module: |
219 | differences between the OpenCL C API and this module: |
194 | |
220 | |
195 | =over 4 |
221 | =over 4 |
196 | |
222 | |
… | |
… | |
201 | =item * OpenCL uses CamelCase for function names (C<clGetPlatformInfo>), |
227 | =item * OpenCL uses CamelCase for function names (C<clGetPlatformInfo>), |
202 | while this module uses underscores as word separator and often leaves out |
228 | while this module uses underscores as word separator and often leaves out |
203 | prefixes (C<< $platform->info >>). |
229 | prefixes (C<< $platform->info >>). |
204 | |
230 | |
205 | =item * OpenCL often specifies fixed vector function arguments as short |
231 | =item * OpenCL often specifies fixed vector function arguments as short |
206 | arrays (C<size_t origin[3]>), while this module explicitly expects the |
232 | arrays (C<$origin[3]>), while this module explicitly expects the |
207 | components as separate arguments- |
233 | components as separate arguments- |
|
|
234 | |
|
|
235 | =item * Structures are often specified with their components, and returned |
|
|
236 | as arrayrefs. |
208 | |
237 | |
209 | =item * Where possible, one of the pitch values is calculated from the |
238 | =item * Where possible, one of the pitch values is calculated from the |
210 | perl scalar length and need not be specified. |
239 | perl scalar length and need not be specified. |
211 | |
240 | |
212 | =item * When enqueuing commands, the wait list is specified by adding |
241 | =item * When enqueuing commands, the wait list is specified by adding |
… | |
… | |
245 | |
274 | |
246 | =over 4 |
275 | =over 4 |
247 | |
276 | |
248 | =item $int = OpenCL::errno |
277 | =item $int = OpenCL::errno |
249 | |
278 | |
250 | The last error returned by a function - it's only changed on errors. |
279 | The last error returned by a function - it's only valid after an error occured |
|
|
280 | and before calling another OpenCL function. |
251 | |
281 | |
252 | =item $str = OpenCL::err2str $errval |
282 | =item $str = OpenCL::err2str $errval |
253 | |
283 | |
254 | Comverts an error value into a human readable string. |
284 | Comverts an error value into a human readable string. |
255 | |
285 | |
… | |
… | |
265 | |
295 | |
266 | Returns all available OpenCL::Platform objects. |
296 | Returns all available OpenCL::Platform objects. |
267 | |
297 | |
268 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetPlatformIDs.html> |
298 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetPlatformIDs.html> |
269 | |
299 | |
270 | =item $ctx = OpenCL::context_from_type_simple $type = OpenCL::DEVICE_TYPE_DEFAULT |
300 | =item $ctx = OpenCL::context_from_type $properties, $type = OpenCL::DEVICE_TYPE_DEFAULT, $notify = undef |
271 | |
301 | |
272 | Tries to create a context from a default device and platform - never worked for me. |
302 | Tries to create a context from a default device and platform - never worked for me. |
273 | |
303 | |
274 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateContextFromType.html> |
304 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateContextFromType.html> |
275 | |
305 | |
… | |
… | |
295 | |
325 | |
296 | =item @devices = $platform->devices ($type = OpenCL::DEVICE_TYPE_ALL) |
326 | =item @devices = $platform->devices ($type = OpenCL::DEVICE_TYPE_ALL) |
297 | |
327 | |
298 | Returns a list of matching OpenCL::Device objects. |
328 | Returns a list of matching OpenCL::Device objects. |
299 | |
329 | |
300 | =item $ctx = $platform->context_from_type_simple ($type = OpenCL::DEVICE_TYPE_DEFAULT) |
330 | =item $ctx = $platform->context_from_type ($properties, $type = OpenCL::DEVICE_TYPE_DEFAULT, $notify = undef) |
301 | |
331 | |
302 | Tries to create a context. Never worked for me. |
332 | Tries to create a context. Never worked for me, and you need devices explitly anyway. |
303 | |
333 | |
304 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateContextFromType.html> |
334 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateContextFromType.html> |
305 | |
335 | |
|
|
336 | =item $ctx = $device->context ($properties = undef, @$devices, $notify = undef) |
|
|
337 | |
|
|
338 | Create a new OpenCL::Context object using the given device object(s)- a |
|
|
339 | CL_CONTEXT_PLATFORM property is supplied automatically. |
|
|
340 | |
|
|
341 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateContext.html> |
|
|
342 | |
306 | =back |
343 | =back |
307 | |
344 | |
308 | =head2 THE OpenCL::Device CLASS |
345 | =head2 THE OpenCL::Device CLASS |
309 | |
346 | |
310 | =over 4 |
347 | =over 4 |
… | |
… | |
312 | =item $packed_value = $device->info ($name) |
349 | =item $packed_value = $device->info ($name) |
313 | |
350 | |
314 | See C<< $platform->info >> for details. |
351 | See C<< $platform->info >> for details. |
315 | |
352 | |
316 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetDeviceInfo.html> |
353 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetDeviceInfo.html> |
317 | |
|
|
318 | =item $ctx = $device->context_simple |
|
|
319 | |
|
|
320 | Convenience function to create a new OpenCL::Context object. |
|
|
321 | |
|
|
322 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateContext.html> |
|
|
323 | |
354 | |
324 | =back |
355 | =back |
325 | |
356 | |
326 | =head2 THE OpenCL::Context CLASS |
357 | =head2 THE OpenCL::Context CLASS |
327 | |
358 | |
… | |
… | |
436 | |
467 | |
437 | =item $ev = $queue->enqueue_write_image ($src, $blocking, $x, $y, $z, $width, $height, $depth, $row_pitch, $data, $wait_events...) |
468 | =item $ev = $queue->enqueue_write_image ($src, $blocking, $x, $y, $z, $width, $height, $depth, $row_pitch, $data, $wait_events...) |
438 | |
469 | |
439 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueWriteImage.html> |
470 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueWriteImage.html> |
440 | |
471 | |
441 | =item $ev = $queue->enqueue_copy_buffer_rect ($src, $dst, $src_x, $src_y, $src_z, $dst_x, $dst_y, $dst_z, $width, $height, $depth, $src_row_pitch, $src_slice_pitch, 4dst_row_pitch, $dst_slice_pitch, $ait_event...) |
472 | =item $ev = $queue->enqueue_copy_buffer_rect ($src, $dst, $src_x, $src_y, $src_z, $dst_x, $dst_y, $dst_z, $width, $height, $depth, $src_row_pitch, $src_slice_pitch, 4dst_row_pitch, $dst_slice_pitch, $wait_event...) |
442 | |
473 | |
443 | Yeah. |
474 | Yeah. |
444 | |
475 | |
445 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueCopyBufferRect.html> |
476 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueCopyBufferRect.html> |
446 | |
477 | |
447 | =item $ev = $queue->enqueue_copy_buffer_to_image (OpenCL::Buffer src, OpenCL::Image dst, size_t src_offset, size_t dst_x, size_t dst_y, size_t dst_z, size_t width, size_t height, size_t depth, ...) |
478 | =item $ev = $queue->enqueue_copy_buffer_to_image ($src_buffer, $dst_image, $src_offset, $dst_x, $dst_y, $dst_z, $width, $height, $depth, $wait_events...) |
448 | |
479 | |
449 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueCopyBufferToImage.html>. |
480 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueCopyBufferToImage.html>. |
450 | |
481 | |
451 | =item $ev = $queue->enqueue_copy_image (OpenCL::Image src, OpenCL::Buffer dst, size_t src_x, size_t src_y, size_t src_z, size_t dst_x, size_t dst_y, size_t dst_z, size_t width, size_t height, size_t depth, ...) |
482 | =item $ev = $queue->enqueue_copy_image ($src_image, $dst_image, $src_x, $src_y, $src_z, $dst_x, $dst_y, $dst_z, $width, $height, $depth, $wait_events...) |
452 | |
483 | |
453 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueCopyImage.html> |
484 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueCopyImage.html> |
454 | |
485 | |
455 | =item $ev = $queue->enqueue_copy_image_to_buffer (OpenCL::Image src, OpenCL::Buffer dst, size_t src_x, size_t src_y, size_t src_z, size_t width, size_t height, size_t depth, size_t dst_offset, ...) |
486 | =item $ev = $queue->enqueue_copy_image_to_buffer ($src_image, $dst_image, $src_x, $src_y, $src_z, $width, $height, $depth, $dst_offset, $wait_events...) |
456 | |
487 | |
457 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueCopyImageToBuffer.html> |
488 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueCopyImageToBuffer.html> |
458 | |
489 | |
459 | =item $ev = $queue->enqueue_task ($kernel, $wait_events...) |
490 | =item $ev = $queue->enqueue_task ($kernel, $wait_events...) |
460 | |
491 | |
… | |
… | |
625 | package OpenCL; |
656 | package OpenCL; |
626 | |
657 | |
627 | use common::sense; |
658 | use common::sense; |
628 | |
659 | |
629 | BEGIN { |
660 | BEGIN { |
630 | our $VERSION = '0.03'; |
661 | our $VERSION = '0.14'; |
631 | |
662 | |
632 | require XSLoader; |
663 | require XSLoader; |
633 | XSLoader::load (__PACKAGE__, $VERSION); |
664 | XSLoader::load (__PACKAGE__, $VERSION); |
634 | |
665 | |
635 | @OpenCL::Buffer::ISA = |
666 | @OpenCL::Buffer::ISA = |